An investigation of Turkish speakers' of English and native English speakers' recognition of the writing of Turkish speakers of English and native English speakers

·_■'· J. X lX i V 7 -i:'-i. ^ '..J '

;' 'f ':~.'г··-' *:·"

/ / 2 . B

AN INVESTIGATION OF TURKISH SPEAKERS' OF ENGLISH AND NATIVE ENGLISH SPEAKERS' RECOGNITION OF THE WRITING OF TURKISH SPEAKERS

OF ENGLISH AND NATIVE ENGLISH SPEAKERS

A THESIS

SUBMITTED TO THE FACULTY OF HUMANITIES AND LETTERS AND THE INSTITUTE OF ECONOMICS AND SOCIAL SCIENCES

OF BILKENT UNIVERSITY

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS

IN THE TEACHING OF ENGLISH AS A FOREIGN LANGUAGE

larcfindan bc^i^lanmi§tir.

BY

PAMELA HOLMES AUGUST 1993

H

н г і ' П

ABSTRACT

Title: An investigation of Turkish speakers' of English and native English speakers’ recognition of the writing of Turkish speakers of English and native English speakers. Author: Pamela Holmes

Thesis Chairperson: Dr. Linda Laube, Bilkent University, MA TEFL Program

Thesis Committee Members: Dr. Ruth A. Yontz, Ms. Patricia Brenner, Bilkent University, MA TEFL Program

This study investigated whether or not Turkish non-native English speakers and native English speakers would recognize the writing of advanced Turkish speakers of English and the writing of native English speakers. It was hypothesized that there were features of the non-natives' writing that made it recognizable as non-native; features which gave the Turkish non-native

English speakers' writing the quality of having a "written foreign accent."

The study attempted to determine if both Turkish non-native speakers of English and native English speakers would perceive differences in the writing of these two linguistic groups and be able to distinguish between non-native and native writing. It was also hypothesized that native English speakers would make the distinction more frequently and more readily than Turkish non-native English speakers, because of a native English

speaker's intuitive sense about what "sounds like English." Essays were collected from Turkish non-native English

speakers and native English speakers. The essays were then read by both non-natives and natives, who were asked to indicate

whether they thought the essays were written by Turkish non­ native English speakers or by native English speakers. The readers were also asked to indicate the reasons for their decisions.

The study provided data on some of the features of non­ native and native writing that the readers used to distinguish between the writing from these two linguistic groups. A serious linguistic analysis of these perceived differences was not a part of this study since the aim of the study was merely to determine if readers were able to recognize non-native and native writing.

Following this, the data showed that these perceived

differences were recognizable to natives and to non-natives as well. This data proved the second hypothesis wrong: natives were not able to recognize Turkish non-native English speakers' writing more frequently and more readily than non-natives could do so.

I V

BILKENT UNIVERSITY

INSTITUTE OF ECONOMICS AND SOCIAL SCIENCES MA THESIS EXAMINATION RESULT FORM

August 31, 1993

The examining committee appointed by the

Institute of Economics and Social Sciences for the thesis examination of the MA TEFL student

Pamela Holmes

has read the thesis of the student. The committee has decided that the thesis

of the student is satisfactory.

Thesis Title

Thesis Advisor

Committee Members

: An investigation of Turkish speakers' of English and native English speakers' recognition of the writing of Turkish speakers of English and native English speakers.

: Dr. Linda Laube

Bilkent University, MA TEFL Program

: Dr. Ruth A. Yontz

Bilkent University, MA TEFL Program

Ms. Patricia Brenner

Bilkent University, MA TEFL Program

We certify that we have read this thesis and that in our combined opinion it is fully adequate, in scope and in quality, as a thesis for the degree of Master of A r t s . Linda Laube (Advisor) Ruth A (Committe

u n \ / i U . U i k O (

Approved for the

Institute of Economics and Social Sciences

V I

ACKNOWLEDGEMENTS

I wish to express my deep gratitude and warm affection to my thesis advisor. Dr. Linda Laube, and to thé other members of my thesis committee. Dr. Ruth A. Yontz, and Ms. Patricia Brenner, without whose efforts this thesis would not exist.

I thank my new friends and classmates at Bilkent University, for making this a joyful experience. For their true friendship and support, I thank Şehnaz and Alev. And, of course. Mutlu for his companionship.

I thank Barbara for always being there for me, and Geri for leading the way.

For their love and patience, I thank my children, Tyler and Patricia, Jason, and Aric.

This thesis is dedicated to the memory of Hildegard Holmes who many years ago gave me a dream, and to the memory of my parents, especially my beloved Mother, who understood all of my dreams.

TABLE OF CONTENTS LIST OF TABLES V I I VIII CHAPTER 1 INTRODUCTION ... 1 Background of the P r o b l e m ... 1 Purpose of the S t u d y ... 2 Research Questions ... 3

Limitations and Delimitations of the Study ... 3

Conceptual Definitions of T e r m s ...5

CHAPTER 2 REVIEW OF LITERATURE ... 6

Introduction ... 6

History of Contrastive R h e t o r i c ...6

Theoretical Background of Contrastive Rhetoric . . . 7

Language Transfer ... 7

Cultural V a r i a t i o n ... 8

Text-type ... 9

Current Research in Contrastive Rhetoric ... 9

Research Based on Reader R e s p o n s e ... 11

C o n c l u s i o n ... 13

CHAPTER 3 RESEARCH M E T H O D O L O G Y ... 14

I n t r o d u c t i o n ... 14

M o d e l ... 14

Sources of D a t a ...14

Non-native and Native Writing Subjects . . . . 15

Non-native and Native Reading Subjects . . . . 15

M a t e r i a l s ... 16 P r o c e d u r e s ... 17 W r i t e r s ... 17 Preparation of E s s a y s ... 18 R e a d e r s ... 18 Conclusion ... 19 CHAPTER 4 ANALYSIS OF D A T A ...20 Identification of E s s a y s ... 20

Differences in Non-native and Native Writing . . . .22

CHAPTER 5 C O N C L U S I O N S ... 24

Review of the S t u d y ... 24

Answers to Research Questions ... 24

I m p l i c a t i o n s ... 26 Final C o n c l u s i o n s ... 29 BIBLIOGRAPHY .31 A P P E N D I C E S ... 34 Appendix A Appendix B Appendix C Appendix D Appendix E Appendix F Appendix G Appendix H Appendix I Appendix J Writers' Consent F o r m ... 34 Readers' Consent F o r m ... 35

Questionnaire for Essay Writers . . . .36

Writers' I n s t r u c t i o n s ... 37

Photo/Writing Prompt ... 38

Readers' Packet Instructions ... 39

Readers' Response Form ... 40

Non-Native Readers' R e s p o n s e s ... 41

Native Readers' R e s p o n s e s ... 42

LIST OF TABLES

TABLE PAGE

1 Identification of Essays by Non-native R e a d e r s ... 21 2 Identification of Essays by Native R e a d e r s ... 21

CHAPTER 1 INTRODUCTION Background of the Problem

This research study will investigate the ability of Turkish non-native English speakers (hereafter non-natives) and native English speakers (hereafter natives) to recognize the English writing of advanced Turkish non-natives and that of natives. It will attempt to determine what it is about the writing of non­ natives that makes it seem not quite "English-like," which results in the writing of non-natives having a recognizable foreign accent.

Research in contrastive rhetoric, comparing non-native English writing to native English writing, has been done with several language groups including German, Korean, Thai, Chinese, Arabic, Japanese, and Spanish. Scovel (1981) used both non­ natives and natives to study the ability of these two linguistic groups to recognize non-native and native writing. In his

study, non-native and native readers were asked to distinguish between non-native and native writing on the basis of short written essays. In this regard, the present research study

resembles Scovel's study in that it focuses on the ability of non-natives and natives to recognize non-native writing because of a perceived written foreign accent. Scovel concluded that non-natives wrote in English as well, or nearly as well, as natives, and that the readers were unsuccessful in

distinguishing between the writing of the two groups. This study proposes that there are features which constitute a foreign accent in the writing of advanced Turkish non-natives, which natives would more readily recognize than non-natives would.

The theory that there is a critical period of language learning hypothesizes that after a certain age, complete oral mastery of a second language is impossible (Lenneberg, 1967, as

summarized by Scovel, 1981). Lenneberg states:

Between the ages of two and three years language emerges by an interaction of maturation and self- programmed learning. Between the ages of three and the early teens the possibility for primaiy acquisition continues to be good; the individual appears to be most sensitive to stimuli at this time and to preserve some innate flexibility for the organization of brain func­ tions to carry out the complete integration of sub­

processes necessary for the smooth elaboration of speech and language. After puberty, the ability for self­

organization and adjustment to the physiological demands of verbal behavior quickly declines. The brain behaves as if it had become set in its ways and primary, basic skills not acquired by that time usually remain

deficient for life. (p. 158)

Thus, adult language learners cannot avoid sounding foreign when they speak in their second language, even after careful instruction. Despite this phonological constraint on the ability of adult non-native speakers to sound like native

speakers, there should be no such constraints on their ability to write like native speakers. Yet research has shown that there are differences in the writing of non-native that make it recognizable as foreign to natives.

Purpose of the Study

The study attempted to determine if the Turkish non-natives can distinguish between the writing of non-natives and natives as frequently and as readily as natives. The hypothesis was that natives would be able to identify the writing of non­ natives, whereas the non-natives would not be able to make this

distinction. It was expected that native English speaking readers would have an intuitive sense of what "sounds like English" and that Turkish non-native English speaking readers would not have the intuitive sense to recognize the differences in non-native and native writing.

Research Questions

The goal of this study differed from much of the research in contrastive rhetoric in that its primary aim was only to provide descriptive evidence that a written foreign accent exists and is recognizable in the writing of Turkish non-natives. The study attempted to answer the following research questions:

I. Are native English speakers able to distinguish between the writing of other native English speakers and the writing of advanced Turkish non-native speakers of English more frequently and more readily than Turkish non-natives can make this

distinction?

II. How do native English speakers and non-native English speakers make the distinction between non-native and native writing?

T.imitations and Delimitations of the Study

Certain limitations were realized in the execution of this study. The availability of eligible Turkish non-native and native writers was limited since the study was conducted in Ankara, Turkey. The writers who were available to the

researcher and willing to participate in the study were selected from the students in the MA TEFL Program at Bilkent University, from a small number of other Turkish non-native English speakers and native English speakers from Bilkent, and from the

Turkish/American Association in Ankara.

The number of subjects selected to read and respond to the writing samples was also limited because of the small population of eligible and willing participants available to the researcher

in Ankara. Since the study was attempting to generalize and offer only preliminary information about the recognition of non­ native and native writing, based on intuition rather than an analysis of the writing, the researcher felt that a large number of readers was not necessary for the purposes of the study.

To offset the limitations and to improve the quality of the study, the writing subjects were carefully selected, using

criteria related to their English proficiency. A questionnaire was used to disqualify any non-native participant whose native language was not Turkish, and those who had acquired English as young children in an English speaking country. Non-natives whose native language was something other than Turkish might bring writing conventions from their first language, and any Turkish non-native subjects who had learned English as children would have learned English before the critical age for language learning. In either case, these non-natives would not be

representative of the Turkish non-natives that were desired for the study.

The writing subjects’ self-reporting on a questionnaire regarding their language learning experience, and their anxiety about writing samples for the study may also be seen as

limitations. Self-report is often misleading in that the subjects may not report the information accurately. Anxiety about having their essays used in the study may cause the

writers to be more careful about how they are writing and result in essays which are not truly representative of non-native and native writing.

In an effort to reduce the anxiety of the participants, the subjects were told only that the study focused on writing, not the differnces in non-native and native writing. The collection of writing samples included all members of the MA TEFL Program in order to prevent the anxiety that might arise if only certain

subjects were asked to write. The writing instructions

(Appendix D) were simple and clear in order to give the subjects an unstressful task to complete.

Conceptual Definitions of Terms

The Turkish subjects' English language proficiency was

determined based on their ages when they began to learn English, and on the fact that they were considered advanced enough to be admitted to the MA TEFL Program at Bilkent University, an

English medium program taught by native English speakers. For this study, advanced level simply means competency in the four basic language skills (listening, speaking, reading, and

writing) and knowledge of language components (vocabulary and grammar). More recent definitions of language proficiency, which include sociolinguistic skills (Cummins, 1980) and

communicative competence (Larsen-Freeman, 1981) are not useful here in that the present research deals only with written

CHAPTER 2 REVIEW OF LITERATURE Introduction

The characteristics of non-native English speakers' writing that make it recognizable, almost intuitively by native English speakers, as "having a foreign accent" have been studied by researchers for over 25 years (Leki, 1991). The study of written discourse accent, or first language influences on writing in a second language, has been defined as contrastive rhetoric (Hudelson, 1988). The history of contrastive rhetoric research and its theoretical background, as well as the various methods that have been employed to study the problem of

contrastive rhetoric will be discussed in this chapter. This chapter will then specifically review some of the research conducted in contrastive rhetoric which compares the writing of non-native English speakers from a variety of Lis to the writing of native English speakers, and contributes to the conceptual framework and rationale for the present study: the recognition of the writing of Turkish non-native speakers of English (non-natives) and native speakers of English (natives) by both non-natives and natives.

History of Contrastive Rhetoric

When foreign students began to arrive in the United States in large numbers during the 1960s, language teachers noticed that these students wrote differently in English than native English speaking American students. The writing of the foreign students was not necessarily wrong, but it was different. One of the

first people to examine this phenomenon was Robert Kaplan (1966), as summarized in Leki (1991), who coined the term: contrastive rhetoric. To Kaplan, the differences in writing implied that the international students had ways of organizing and presenting what they wrote that did not match the standards of English, and he sought to determine the nature of those

differences (Purves, 1988). Kaplan's early study of contrastive rhetoric was more intuitive than scientific, but it was valuable in that it established contrastive rhetoric as a new field of inquiry. The early research in contrastive rhetoric was also product-orientated and prescriptive (Leki, 1991).

During the 1970s, discourse analysis focused on spoken rather than written discourse, and did not contribute much to the explanation of the differences in the larger segments of non-native writing (Leki, 1991). This period saw a decline in contrastive rhetoric research, as scholars seem to have lost interest in the subject.

The 1980s brought renewed interest in contrastive rhetoric and in the exploration of more than the surface features of

written discourse. Among the most important research studies of the early 1980s were Hind's (1980, 1984, 1987) studies of

Japanese rhetoric, Purves and Takala's (1982) analysis of essays from students in 14 countries, and the 1984 issue of the

Annual Review of Applied L·inαuisticswhich was devoted entirely to contrastive rhetoric (all summarized by Leki, 1991).

Theoretical Background of Contrastive Rhetoric

In the past decade, most researchers in contrastive rhetoric have recognized that the ability to write a fluent, coherent text implies more than the ability to control vocabulary,

syntax, and mechanics. Theoretical support for a study on the differences in non-native and native writing comes from the current research in contrastive rhetoric which focuses on language transfer, cultural linguistic variations, and text- types .

Language transfer

Whether or not a language learner's linguistic behavior (including writing) can be predicted by the similarities and differences between his native language and the target language

8

is the basis for current research in language transfer. Odlin (1989) points out that the role of language transfer in second language acquisition has always been a controversial topic.

Early contrastive rhetoric research relied heavily on the theory of language transfer, and resulted in a prescriptive application of research findings. Current contrastive research is still attempting to determine when and how language transfer affects second language learning (Larsen-Freeman and Long, 1991).

Cultural variation

Purves (1988) points out that in written English, many academic disciplines require that individuals learn to write according to certain explicit and implicit conventions that affect patterns of organization, syntax, phrasing, and lexicon. If these recognizable differences in writing conventions occur in written English, it is reasonable, Purves feels, to expect that similar kinds of differences separate ethnolinguistic or geographically distant cultures and societies. Purves states that "such is the emergent line of reasoning behind contrastive rhetoric" (p.l4).

Grabe and Kaplan (1989) also believe that the effort to

understand how writing in a second language is influenced by the cultural and linguistic conventions of the writer's first

language continues to provide an impetus for contrastive

rhetoric studies. Grabe and Kaplan maintain that "contrastive rhetoric predicts that writers composing in different languages will produce rhetorically distinct texts, independent of other causal factors such as differences in processing, in age, in relative proficiency, in education, in topic, in task

complexity, or in audience" (p. 264). They point out, though, that this notion does not support the deterministic view that speakers of other languages think differently or have different

cognitive frameworks. They claim that literacy skills are learned and that they are culturally shaped.

Text-type

Grabe (1987) conducted research on English expository prose and found that within this major text genre, a number of text- type distinctions exist. Having reached this conclusion for English expository writing, Grabe poses several questions for other languages: "Does expository prose of some type exist in some other languages? If it does, what are its characteristic properties? What sub-types exist?" (p.ll6) Following his

research on contrastive rhetoric and text-types Grabe concludes that "the basic issue is whether similar text-type

generalizations may be made for different languages" (p.ll6). One problem with text-type comparison is related to the

frequency of a particular text-type in a specific culture. If a text-type appears frequently in one culture, but rarely in

another, the results of a comparative study would be invalid because that particular type of text appears with different frequencies in the two cultures (Leki, 1991).

Current Research in Contrastive Rhetoric

One focus of contrastive rhetoric inquiry concentrates on texts produced by non-native in an effort to: determine various but similar patterns of written discourse among speakers of the same language, contrast these patterns to native writing, and trace these various patterns to patterns of written discourse in the writer's native language (Hudelson, 1988).

Scarcella (1984) examined how non-native writers orient their readers in expository writing. Her results indicate that native writers employ a variety of linguistic devices to engage their readers attention, while non-natives were more limited in their ability to orient their readers. Scarcella feels that this may result from the non-native writ e r s ' unfamiliarity with the norms

10

of expository writing in English. She also says: "They may have been transferring the rules and norms of their own cultures when communicating in English" (p.683). Scarcella also discusses reader/writer familiarity:

Although both natives and non-natives wrote for the same readers,their familiarity with these readers differed

greatly. Native writers shared more knowledge and experience with their readers. As a result, they were

better able to predict their readers' personal

characteristics,including interests, intelligence, and knowledge of the world. Thus, before pen was ever put to paper, natives had a distinct advantage over their non-native counterparts. (p .684)

In their study of cultural variation in reflective writing, Bickner and Peyasantiwong (1988) compared essays written by natives to those written by Thai non-natives. Their research found differences in the way the two groups defined terms and concluded their essays. Also, there were differences in the conversational tone and in vocabulary: formal vs informal.

Research in contrastive rhetoric which focused on persuasive writing was conducted by Connor and Lauer (1988). Their

subjects were all native from the United States, England, and New Zealand. Using an analytic scale to evaluate the samples, the researchers were able to identify differences in the use of credibility and affective appeals in the persuasive writing of their subjects. This cross-cultural research indicates that differences exist even in the writing of native speakers from similar but different cultures and educational backgrounds.

McKay (1989) explores the area of written discourse accent in terms of topic development. She examines two groups of

essays: one written by Chinese students of English in China and the other by students from various ethnic groups studying

11

English in the United States. She states that her purpose w a s : "to share my findings regarding the predictability of how the two groups of students develop the identical topic in different ways based on their cultural and social experiences" (p. 253). Her study was aimed at discovering if culturally determined rhetorical patterns are present in these two groups of essays. Among the patterns that McKay identifies in her examination of the essays Chinese students' are the presentation of a moral lesson, description, an element of time, and reference to public opinion. McKay concludes by stating that: "The question of

just what part social and cultural experiences play in a

writer's development of a topic is an area that needs further research" (p. 261). The significance of McKay's research is that it demonstrates clearly that what students write is influenced by their cultural, social, and educational

experiences. In her study of topic development in non-native writing, McKay concludes that teachers should select topics that are within the realm of the students' experiences, and then

strive to become aware of our own cultural expectations. Research Based on Reader Response

Some researchers in contrastive rhetoric have studied the reactions to non-native writing by both non-natives and natives to determine what differences occur in non-native writing that make it recognizable to native readers. Scovel (1988) asked non-natives and natives to read written essays and listen to recorded speech samples of non-natives and natives. His

readers/listeners were asked to determine if the samples were written/spoken by Americans. Most judges, both non-natives and natives were able to detect the foreign accent easily in the oral samples, but the written essays were more difficult to identify correctly. In Scovel's study, most of the non-native writers were able to "deceive" the judges, and their writing was

12

passed as native by both the non-native and native judges. He concluded that writing in a foreign language can be mastered fully, but not speaking without a foreign accent because of the biological constraints on the acquisition of a second language. The expected results of the current study are opposite those found by Scovel. This researcher expects that native readers will be able to recognize a foreign accent in non-native writing and will intuitively sense something "not quite like English" about the non-native essays.

Santos (1988) took another direction in his research when he studied the reactions of both non-native and native professors to non-native writing. Essays written by non-natives were

evaluated for language and for content. Santos found that while most of the professors were willing to look beyond deficiencies in language to the content of the non-native essays, the non­ native professors rated the language as less acceptable than the native professors did. This is significant since the present study will include both native and non-native readers.

Another research study that examines the reactions of readers to non-native writing was conducted recently by Kobayashi (1992). In this study, natives and Japanese non­

natives were asked to evaluate Japanese non-native compositions in terms of grammaticallity, clarity of meaning, naturalness, and organization. The results of Kobayashi's study support his belief that natives can recognize more readily the differences in the writing of non-natives and natives. They are able, he feels, "...to judge the acceptability of certain expressions by intuition, drawing upon implicit knowledge" (p.82). He further states that:

It is, after all, the native speaker who is the authority in this matter....It is sometimes said, for example, that the best definition of a dog

13

is that it is an animal that is recognized as a dog by other dogs. (p. 105)

Conclusion

Since Kaplan first became interested in the writing of foreign students in the 1960's, various research studies have been conducted in the field of contrastive rhetoric to determine the causes for the differences in non-native and native writing. It is generally accepted that the writing of non-natives will be different than the writing of natives, and recent research has concentrated on language transfer, text-types, and cultural variations as some of the possible influences on non-native writing. At least one researcher, Scovel, maintains that

advanced non-native writing should be indistinguishable from the writing of natives, as confirmed by his 1981 study.

The study follows Scovels' model to see if the writing of non-natives is recognizable to non-natives and natives. The study will not attempt analyze the perceived differences or the causes of the differences. Analysis of the differences in the writing of advanced Turkish non-natives and that of natives will await further research in contrastive rhetoric.

CHAPTER 3 RESEARCH METHODOLOGY Introduction

The purpose of this study was to determine (1) if Turkish non-native English speakers (non-natives) were able to

distinguish between the writing of non-natives and native

English speakers (natives) as readily as natives can make this distinction and (2) what were the characteristics in the writing of advanced Turkish non-natives that made it recognizable as

"having a foreign accent." In order to answer these questions, writing samples from both Turkish non-natives and natives were collected and read by both non-natives and natives.

This chapter describes the data collection procedures, the materials, and the participants used for this descriptive study on non-natives' and natives' writing.

Model

The present study most closely follows Scovel's (1981) study in which ten highly competent non-natives and ten natives were asked to write essays which were then presented to adult native judges. The judges were posed with only a simple task: separate the essays written by natives from those written by non-natives. However, in the present study, both native and non-native readers were asked to distinguish native from non­ native writing, and they were asked to indicate the reasons for their judgements.

Sources of Data

Four separate groups of participants were needed to conduct the present study. The first two groups consisted of the essay writers: ten advanced Turkish speakers of English and ten native English speakers. The other two groups were the essay readers: five Turkish non-natives and five natives.

15

Non-native and Native Writing Subjects

The Turkish non-native essay-writing subjects were selected from the class members of the MA TEFL Program at Bilkent

University in Ankara. The students in this program were all English language teachers at their home universities in Turkey, Azerbaijan or Kazakhstan. The class consisted of twenty-four potential subjects (excluding the researcher) who were

considered to be at an advanced level in English based on their acceptance into the MA TEFL Program, an English-medium program taught by four native American Fulbright professors, and on

their scores of seventy-three or higher on the Michigan Language Proficiency Test.

Subjects for native writing samples were chosen from the English language teaching staff at the Turkish/American Association in Ankara. The group of American native English

speaking volunteers consisted of three male and seven female essay writers, whose ages ranged from 22 years to 40 years.

Their English language teaching experience was varied, with some having as much as thirteen years' experience, while others were in their first year of teaching English.

Non-native and Native Reading Subjects

Four of the Turkish non-native participants who were asked to read the essays were also chosen from the MA TEFL Program at Bilkent University. Only subjects who qualified, based on the information retrieved in the questionnaire were asked to read the essays. One non-native reader was from the Faculty of Business at Bilkent University, himself a visiting professor from the United States. The non-native readers included three males and two females.

One native reader was a faculty member of Bilkent

University's School of English Language (B.U.S.E.L.). Another was a teacher in Freshman English at Bilkent. The other three

16

native readers were from the English language teaching staff at the Turkish/American Association in Ankara.

Materials

All potential candidates for the essay writing task and the essay reading task were given consent forms (Appendices A and B ). These forms explained what would be required of the

participants, gave a brief description of researcher's goals, and assured the participants of the confidentiality of the study.

The non-natives who consented to write sample essays for the study were presented with a questionnaire (Appendix C). The purpose of this questionnaire was to discover any non-native essay writers whose native language was not Turkish. Since the study focused on the writing of only Turkish non-natives,

possible subjects whose native language was other than Turkish needed to be identified and disqualified from the study. The questionnaire also sought to determine the length of time that each subject had been speaking and teaching English, since it was believed that those who had the greatest experience with English would be more able to write like natives and thus be more valid subjects for the study.

Both groups of sample essay writers were given the same writing instructions and picture prompt (Appendix D and E), and were provided with paper on which to write their essays. Each participant was assigned a code number in order to keep the identities of the participants confidential, but which could be used for identification purposes by the researcher. The

instructions asked that the essay writers use only their

assigned numbers to label their work, and that they write for approximately thirty minutes.

The picture prompt was chosen because of its potential emotional impact and because it was culturally unbiased. The

photograph showed a person in a wheelchair near a small tree at the edge of a lake. The essay writers were asked to write a short composition based on their responses to the prompt, either a fictional narrative or an essay based on their own personal experiences.

Procedures Writers

All MA TEFL class members were presented with the consent form during a regular class meeting at Bilkent University. Eighteen members of the class of twenty-four volunteered to participate in the essay writing task. The questionnaire,

writing instructions, and picture prompt were then presented to these volunteers. Based on the participants' responses to the questions on the questionnaire, any subject whose native

language was not Turkish was eliminated from the study. Of the remaining Turkish non-natives, those who had been teaching

English for the greatest length of time were chosen over those with less English teaching experience. The essays of ten

participants were chosen as suitable examples of non-native writing, based on the fact that the topics in their writing did not reflect their nationality or culture. The final non-native subjects who were chosen for the study ranged in age from 25 to 39, and the group included six females and four males.

Ten American members of the English language teaching staff at the Turkish/American Association were chosen as subjects to write native sample essays for the study, based solely on their willingness to cooperate. Because of a conflict with the

participants' teaching schedules, the samples were collected at two different times, with five subjects present each time, in a classroom at the Turkish/American Association. Attempts were made to recreate the same conditions for collecting the samples each time, and the same materials were used. Each subject was

18

given a consent form, writing instructions with the picture prompt, and paper. The volunteer essay writers from

Turkish/American Association were not asked to complete the questionnaire since they were all native English speakers employed as English language teachers in Ankara. The native subjects ranged in age from 22 to 40, and included seven females and three males.

Preparation of Essays

The ten sample essays collected from the Turkish non-native writers from the M A TEFL program at Bilkent and the ten sample essays collected from the American teachers of English from the Turkish/American Association were re-typed for the sake of

conformity and ease of reading. Any references to geographic locations and any names in both the non-native and native essays which might indicate the writer's native language or culture were removed from the essays and replaced with XXXX. The essays of the non-natives and the natives were then combined in a

random order for presentation to the readers. Readers

Each of the ten readers (five non-native and five native) was then asked to read all 20 essays. The essay packet

instruction form (Appendix F), asked the readers to separate the essays they thought were written by non-natives from those

written by natives. The purpose of the study was to determine the ability of non-native and native readers to distinguish native from non-native writing, and not to analyze any

differences that were perceptible to the readers. However, the readers were asked to try to identify the reasons for their

judgments and to write in their own words on the reader response form (Appendix G) why they judged each essay to be written by a non-native or a native, or to indicate on the essays themselves, what it was that sounded "not like English."

19

Conclusion

The results of the identification task given to the readers were divided into two categories: the responses of the native readers to the essays and the responses of the non-native

readers to the essays. The ability of each Turkish non-native and each native reader to separate native writing from non­ native writing was tallied, and percentages were established which indicated how frequently each of the two groups of readers were successful in their ability to distinguish between essays written by non-natives and those written by natives. The reasons that each reader gave for deciding whether each essay was written by a non-native or a native were also categorized and tallied.

CHAPTER 4 ANALYSIS OF DATA Identification of Essays

The ten Turkish non-native speakers' (non-native) essays were read by ten readers (five non-natives and five natives), making a combined total of 100 readings. Out of those 100 readings, the non-native essays were correctly identified 90 times, with only ten readers incorrectly identifying non-native writing as native writing.

The native English speakers' (native) essays were also read a total of 100 times. There were 75 correct identifications of native writing, with 25 incorrect identifications of native writing as non-native writing (Appendices H and I).

These results are taken from the combined responses of both groups of readers. When the data is analyzed further,

separating the responses of the non-native readers and the native readers, it becomes apparent that the ability of non­ natives to recognize non-native writing is very close to that of the natives' ability do so.

Table 1 shows that the non-native readers were able to

identify non-native essays 86% of the tjLme. Table 2 shows that the native readers were able to identify non-native essays 94% of the time. The tables also show that both groups were less accurate in the identification of native writing. Both the non­ native and the native readers agreed in the identification only 74% of the time.

Since there is not a remarkable difference in the abilities of the two groups to recognize non-native and native writing, the data do not support the hypothesis that natives are able to distinguish between the writing of other native English speakers and the writing of advanced Turkish non-native English speakers more frequently and more readily than non-natives can make this distinction.

Identification of Essays bv Non-native Readers

Table 1

21

Identification of Essays by Native Readers

Native Readers Total Correct Identification of Non-native and Native Essays Correct Identification of Non-native Essays Only Correct Identification of Native Essays Only 5 Readers x 20 Essays = 84% 84/100 = 94% 9.4/10 = 74% 7.4/10

There was a close agreement between the non-native and native readers on the correct identification of combined essays, with non-natives being correct 80% of the time, to the natives' 84%.

On the recognition of non-native writing, the non-natives were correct 86%, while the natives were 94% correct. Both groups were less able to recognize native writing, with each being correct only 74% of the time.

Differences in Non-native and Native Writing

The non-native and native readers were asked to write in

their own words on the response form, or to indicate on the

essays themselves, what it was about the essays that made them

decide whether each essay was written by a non-native or a

native.

The most frequently mentioned characteristics of the

correctly identified non-native essays that both non-native and

native readers gave as indicators that the writing was non­

native English are listed and counted below:

Distinguishable features:

Natives

Non-nati

Article use (15)

(10)

(5)

Awkward sentence structure (12)

(6)

(6)

Word choice or vocabulary (11)

(3)

(8)

Verb tense and aspect (10)

(5)

(5)

Sounds as if translated (10)

(6)

(4)

Lack of flow; hard to follow (8)

(6)

(2)

Language is too formal (5)

(3)

(2)

Prepositions (5)

(4)

(1)

Omission of "and" in a series (4)

(3)

(1)

The most frequently mentioned characteristics of the

correctly identified native essays that both non-native and

native readers gave as indicators that the writing was native

English are listed and counted below:

Distinguishable features:

Natives

Vocabulary (26)

(5)

Native-like expressions (11)

(2)

Makes sense; "flows" (11)

(9)

Non-native (2 1 )

(9)

"Sounds" like native writing (10) Writing style (8)

Sentence structure (7)

Knowledge of the language (4)

(

1

2

(6)

1

23

These responses show how the readers differentiated between non-native and native writing. They indicate that the non­

native readers identified the same characteristics in the sample essays as the native readers. None of the readers mentioned rhetorical features in the essays as the reasons for their judgments. Only lexical and grammatical features were identified.

24

CHAPTER 5 CONCLUSIONS Review of the Study

Most research in contrastive rhetoric confirms that non­

natives write differently in English than natives write in their native language. These differences in non-native writing,

regardless of their cause (i .e l a n g u a g e transfer, text-type, or cultural linguistic variations), are what give non-native

writing a recognizable "foreign accent." However, Scovel's (1981) study resulted in finding that the writing of advanced non-natives was indistinguishable from the writing of natives. In order to challenge Scovel's results, this study first proved the writing of advanced Turkish non-natives and the writing of natives is in fact distinguishable.

This study examined the ability of readers from the same two linguistic groups to distinguish between the writing of non­ natives and natives. It was hypothesized that natives would recognize non-native writing more frequently and more readily than non-natives would be able to do so.

The study also asked the non-native and native readers to indicate what it was about the writing samples that made it seem like "native" or "non-native" writing. Their responses were not used in an attempt to explain or examine the causes of the

perceived differences, but merely to determine what

characteristics of the writing samples led the readers to their conclusions.

Answers to the Research Questions

The answer to the first research question was negative, disproving the hypothesis that native English speakers could recognize non-native writing more readily than non-native

English speakers could do so. Only by a small percentage, 94% vs. 86%, were natives more accurate in distinguishing between non-native and native writing. This result was of interest to the researcher, not only because it proved the hypothesis to be wrong, but because it showed that non-natives were able to

differentiate between the writing of non-natives and natives without the intuitive sense that only natives were expected to have.

Responses to the essays from the native and non-native

readers provided evidence that differences were perceived in the writing from these two linguistic groups. Again, without giving detailed descriptions of linguistic features, but rather their intuitive reactions to the essays, both groups were able to indicate specifically which items in the essays caused them to decide if the writing was non-native or native. Both non­

natives and natives gave the same reasons for their judgments. Here, too, the researcher found it interesting that the non­ native readers identified the same features in the writing samples that made it seem "like English" or "not like English."

By showing that readers are able recognize non-native writing, the data supports previous research in contrastive rhetoric which accepts that there will be perceptible

differences in non-native and native writing. However, the data do not confirm Kobayashi's (1992) study in which native speakers were able to recognize non-native writing more readily than non­ natives, based on some intuitive sense of what sounds "like English." The analysis of the data collected in this study indicates that non-natives as well as natives were able to recognize non-native writing.

The data which was examined in this study results in a conclusion which is the opposite of that found in Scovel's

(1981) study in which non-native writers were able to "deceive" readers with their "native-like" writing. In the present study, non-native writing was recognized by both non-native and native readers.

Implications

While this study does not claim to make new pedagological implications for teaching ESL/EFL, it supports the implications of other research in contrastive rhetoric aimed mostly at the need for language teachers to be aware of the differences they will see in their students’ writing.

Kaplan (1988) summarizes ten objectives based on cultural and linguistic differences:

(1) To make the teacher of composition aware that: (a) Different composing conventions do exist in

different cultures and that these different conventions need to be addressed in teaching composition; the fact that a student is able to compose in one language does not mean that the student can compose in any other language.

(b) Certain grammatical features function at the level of discourse; the fact that a student has control only of certain sentential syntax does not mean the student can generate text.

(c) There is a relative distribution of reader/writer responsibility in different cultural systems and that this distribution affects assumptions about audience and about shared knowledge; the fact that a student

understands audience in one language does not mean the student understands audience in any other language system.

(d) A composition is a product arrived at through a process; the fact that the student may generate one relatively successful text does not mean the student understands the process.

(e) Morphosyntactic competence is prerequisite to writing; the fact that a student can understand the

structure of individual sentences in a language does not mean that the student can understand the structure of cohesion and coherence in text in that language.

(2) To make the student of composition aware that:

(a) Audience must be defined before composition can be undertaken; the assumption that the audience is the writer is not always valid.

(b) There are a number of different writing acts -

writing without composing, writing through composing in which the essential purpose is reporting; and writing through composing in which the composing process

functions as a heuristic act - and that each of these requires different strategies; the fact

that a student is capable of writing in one language does not mean the student is aware of the frequency and distribution of writing tasks in any other language. (c) There exists a set of text conventions that the student is expected to be able to manage; the fact that the student knows the conventions of his or her own writing system does not mean the student understands the

conventions employed in the target language.

(d) In order to compose anything, the student must have, and be able to bring to bear on the composing task, a universe of knowledge, including both "world knowledge" and technical knowledge of the subject; the fact that a student has opinions about a given topic doe not mean the student knows the subject well enough to write about it.

(e) Writing is a social phenomenon - a technique for negotiating meaning with other identifiable set of human beings - which requires far more than a minimal control of syntactic and lexical items in the target language.

(p.296,297)

Other researchers whose studies in contrastive rhetoric were focused on language transfer, text-type, or cultural linguistic variations have pointed out additional pedagological

implications when concluding their studies.

According to Odlin (1989), one of the reasons for language teachers to consider more closely the problem of language

transfer is that teaching may be more effective with an

understanding of the differences between languages. Odlin also feels it is important for teachers to communicate the message to students that their linguistic and cultural background is

respected by the teacher.

Kobayashi (1992) states: "The pedagogical goal of writing is twofold: to have learners create a final product that is logical, persuasive, and error-free, and to train them to be good negotiators with their own ideas. Demanding of learners only the first goal will engender frustration and even the loss

of confidence, just as does demanding perfect native-like English pronunciation." (p.l07)

Scarcella (1984) points out that teachers with an awareness of cultural linguistic differences will be able to set more reasonable goals for their students, rather than expecting them to produce something which they cannot. Scarcella further

suggests: "To develop discourse knowledge, teachers might expose students to the types of writing they would like their students to compose. To develop cultural knowledge, teachers might give their students information on topics which are the focus of writing assignments" (p.684).

Language teachers should broaden their concept of what

constitutes "good work" and not require their students to share and reproduce in their writing our world view, one to which they are, of course, alien (Land and Whitley, 1989).

Final Conclusions

Determining specifically which aspects of non-native writing contribute to a written foreign accent may be difficult since the non-native may lack proficiency in several areas, including grammatical accuracy, lexical selection, and rhetorical

organization. While the present study did not provide

linguistic analysis of non-natives writing, it did show that readers are able to discern between the writing of advanced Turkish non-natives and natives.

The scope of the present study was small, focusing on the ability of readers to recognize non-native writing. Because of the limitations of the study, only a small amount of data was collected. More comprehensive future studies could attempt to determine if differences in rhetorical organization exist in

Turkish non-native writing. Further research might try to discover if advanced non-native English speakers develop a native-like intuitive sense of what sounds "like English" enabling them to recognize non-native and native writing.

Future studies in contrastive rhetoric might determine if the syntactical, lexical, and grammatical features which identified the writing of advanced Turkish non-native English speakers are the result of language transfer, text-type, cultural linguistic variation, or a combination of these and other possible factors, including lack of proficiency.

BIBLIOGRAPHY ^ 1 Bickner, R., & Peyasantiwong, P. (1988). Cultural variation in

reflective writing. In A. Purves (Ed.), Writing across languages and cultures; Issues in contrastive rhetoricfpp. 160-174). Newbury Park, CA: Sage.

Connor, U . , & Kaplan, R. (1987). Introduction. In U. Connor, & R. Kaplan (Eds.), Writing across languages: Analysis of L2 text (pp. 1-5). Reading, MA: Addison-Wesley.

Connor, U., & Lauer, J. (1988). Cross-cultural variation in persuasive student writing. In A. Purves (Ed.), writing across languages and cultures; Issues in contrastive rhetoric (pp. 138-159). Newbury Park, CA; Sage.

Cummins, J. (1980). The cross-lingual dimensions of language proficiency; Implications for bilingual education and the optimal age issue. TESOL Quarterly. 1 4 . 175-188.

Grabe, W. (1987). Contrastive rhetoric and text-type research. In U. Connor, & R. Kaplan (Eds.), Writing across languages; Analysis of L2 Text (p p. 115-137). Menlo Park, CA; Addison- Wesley.

Grabe, W . , & Kaplan, R. (1989). Writing in a second language; Contrastive rhetoric. In D. Johnson, & D. Roen (Eds.),

Richness in writing; Empowering ESL students (pp. 263-283). NY; Longman.

Hudelson, S. (1988). Writing in a second language. In Kaplan R. (Ed.), Annual review of applied linguistics. 9 (pp. 210- 222). NY; Cambridge University Press.

Kaplan, R. (1988). Contrastive rhetoric and second language learning: Notes toward a theory of contrastive rhetoric. In A. Purves (Ed.), Writing across languages and cultures;

Issues in contrastive rhetoric (pp. 275-304). Newbury Park, CA; Sage.

Kobayashi, T. (1992). Native and nonnative reactions to ESL compositions. TESOL Quarterly. 26 (1), 81-112.

Land, Jr., R . , & Whitley, C. (1989). Evaluating second language essays in regular composition classes: Toward a pluralistic U.S. rhetoric. In D. Johnson, & D. Roen (Eds.), Richness in writing: Empowering ESI students (pp. 285-293). NY: Longman.

Larsen-Freeman, D. (1981). The 'what' of second language

acquisition. In Hines, M., & Rutherford, W. (Eds.), On TESOL '81 (pp. 107-28). Washington, D.C.: TESOL.

Larsen-Freeman, D . , & Long, M. (1991). An introduction to second language acquisition research. London: Longman Leki, I. (1991). Twenty-five years of contrastive rhetoric.

TESOL Quarterly. 25 (1), 123-143.

McKay, S. L. (1989). Topic development and written discourse accent. In D. Johnson, & D. Roen (Eds.), Richness in

writing: Empowering ESL students (pp. 253-262). NY: Longman.

Odlin, T. (1989). Language transfer: Cross-linguistic

influence in language l e a r n i n g . NY: Cambridge University Press.

Purves, A. (1988). Writing across languages and cultures. In A. Purves (Ed.), Issues in contrastive rhetoric (pp. 9-21). Newbury Park, CA; Sage.

Santos, T. (1988). Professors' reactions to the academic

writing of nonnative-speaking students. TESOL Quarterly. 22 (1), 69-91.

Scarcella, R. (1984). How writers orient their readers in expository essays: A comparative study of native and non­ native English writers. TESOL· Quarterly. 18 (4), 671-687. Scovel, T. (1981). The recognition of foreign accents in

English and its implications for psycholinguistic theories of language acquisition. In J-G.Savard, & L·. Eaforge (Eds.), Proceedings of the 5th Congress of AIEA (pp.389-401).

Eaval: University of Eaval Press.